sentence processing conference in NYC
This is my last day in New York -- this evening I get on a plane to Seattle to see everyone I know who lives on the West and who I could trick into coming to Seattle to visit me.
NYC is really great, but I'm not sure I could live here. Despite having lived in cities for the past ten years, I think I must have somehow been marked psychologically by the years I spent in flat, wide-open places and in places where there were undeveloped spaces larger than I could explore on foot within easy walking distance. This environment is really exciting, but it doesn't take long for me to start feeling totally overstimulated.
If you're sort of used to seeing me around and noticed that I haven't been, it's because I've been feverishly preparing for this conference. Now I get my life back!
I had an abstract for a poster accepted with a postdoc in my lab some months ago. He came up with a model that took a certain kind of syntactic structure and predicted how it would be pronounced. I noticed that it was a parseable model, and told him as much, but he was skeptical, so I wrote the parser. When I ran it on the actual human utterance data it was really successful -- I was quite surprised. (The poster is here -- the first three columns are the other guy's work; the last two are mine.)
You might think that because I was one of the lead authors on that poster that was the one I was preparing for. But no, most of what I was doing was a bunch of data analysis programs for another person in my lab who had a lot of data but no way to analyze it herself. So I wrote many programs. But she promised me that in exchange for this, I would be cut in on more conceptually interesting work in the future. And of course, my name went on the poster. Go me.
I got up at 4:30 on Wednesday and set off to catch the train, and then worked on my poster on the train, and most of the day in NYC until I met my brother to see Avenue Q. Then I worked some more. I stayed at a really nice little hotel called the Hotel Deauville. It was a short walk from the conference site near the Empire State Building, and my room was only $80 a night, because it was on the 7th floor, and the bathroom was on the 3rd. This doesn't really bother me, but apparently for most people it's totally unacceptable. If you're going to stay in a hotel in NYC and you don't mind climbing stairs to use the bathroom I highly recommend it.
In the morning I sent the file off to a Kinko's, and then went there to make sure that everything was okay and that I could pick it up the following morning, then I went to the conference. Its topic is psycholinguistic sentence processing, which is not a topic I'm really passionate about, but my advisor is. So. It's interesting insofar as these guys are actually interested in figuring out how people process language -- they're not just throwing huge, psychologically implausible computational models full of gargantuan matrices at the language problem to get immediate results. But ultimately, sentences aren't really my thing. Fortunately, I can be interested in just about anything for a while.
I left the conference halfway through the reception at 7 pm that evening, went back to my hotel room, and slept for almost 12 hours. This was a good thing -- I don't imagine that my poster presentation would have gone nearly as well if I hadn't done that. And it went very well indeed! Almost everyone said very positive things about the results of my work. As soon as I started presenting the poster I immediately noticed a dozen things that were wrong with it. (We'll just call that a learning experience.) But from the comments that most people made about my fairly simple parser, I'm pretty sure that in this environment I'm in the top decile in my ability to do computational work. When I take into account that I'm still at a point where my programming abilities are improving noticeably from one month to the next, that makes me feel somewhat hopeful. My basic strategy for my career is to find some area of cognitive research where my mathematical and computational abilities will give me a noticeable edge. The past few days have made me feel better about that.
The evening was a bit of a trial. I was already feeling overstimulated from being in the city and from all of the talking to people I'd done during my poster presentation. That alone was probably 10 times as much interaction with people as I'm used to in a day, and more than 100 times the interaction I usually have with strangers. Then Ted invited everyone at the conference who had a significant association with his lab (about 16 people) to dinner at a fancy Italian place on restaurant row. I'm not one to decline a good meal, but I was feeling tired and introverted. The dinner was fantastic, but it lasted three and a half hours! By the time it ended I felt like I was about to suffer a total mental collapse. I was really glad to get back to my room and sleep.
Today I go to the conference until about 3, and then leave for the airport. Next time most of you hear from me I'll probably be in Seattle. Until then!
NYC is really great, but I'm not sure I could live here. Despite having lived in cities for the past ten years, I think I must have somehow been marked psychologically by the years I spent in flat, wide-open places and in places where there were undeveloped spaces larger than I could explore on foot within easy walking distance. This environment is really exciting, but it doesn't take long for me to start feeling totally overstimulated.
If you're sort of used to seeing me around and noticed that I haven't been, it's because I've been feverishly preparing for this conference. Now I get my life back!
I had an abstract for a poster accepted with a postdoc in my lab some months ago. He came up with a model that took a certain kind of syntactic structure and predicted how it would be pronounced. I noticed that it was a parseable model, and told him as much, but he was skeptical, so I wrote the parser. When I ran it on the actual human utterance data it was really successful -- I was quite surprised. (The poster is here -- the first three columns are the other guy's work; the last two are mine.)
You might think that because I was one of the lead authors on that poster that was the one I was preparing for. But no, most of what I was doing was a bunch of data analysis programs for another person in my lab who had a lot of data but no way to analyze it herself. So I wrote many programs. But she promised me that in exchange for this, I would be cut in on more conceptually interesting work in the future. And of course, my name went on the poster. Go me.
I got up at 4:30 on Wednesday and set off to catch the train, and then worked on my poster on the train, and most of the day in NYC until I met my brother to see Avenue Q. Then I worked some more. I stayed at a really nice little hotel called the Hotel Deauville. It was a short walk from the conference site near the Empire State Building, and my room was only $80 a night, because it was on the 7th floor, and the bathroom was on the 3rd. This doesn't really bother me, but apparently for most people it's totally unacceptable. If you're going to stay in a hotel in NYC and you don't mind climbing stairs to use the bathroom I highly recommend it.
In the morning I sent the file off to a Kinko's, and then went there to make sure that everything was okay and that I could pick it up the following morning, then I went to the conference. Its topic is psycholinguistic sentence processing, which is not a topic I'm really passionate about, but my advisor is. So. It's interesting insofar as these guys are actually interested in figuring out how people process language -- they're not just throwing huge, psychologically implausible computational models full of gargantuan matrices at the language problem to get immediate results. But ultimately, sentences aren't really my thing. Fortunately, I can be interested in just about anything for a while.
I left the conference halfway through the reception at 7 pm that evening, went back to my hotel room, and slept for almost 12 hours. This was a good thing -- I don't imagine that my poster presentation would have gone nearly as well if I hadn't done that. And it went very well indeed! Almost everyone said very positive things about the results of my work. As soon as I started presenting the poster I immediately noticed a dozen things that were wrong with it. (We'll just call that a learning experience.) But from the comments that most people made about my fairly simple parser, I'm pretty sure that in this environment I'm in the top decile in my ability to do computational work. When I take into account that I'm still at a point where my programming abilities are improving noticeably from one month to the next, that makes me feel somewhat hopeful. My basic strategy for my career is to find some area of cognitive research where my mathematical and computational abilities will give me a noticeable edge. The past few days have made me feel better about that.
The evening was a bit of a trial. I was already feeling overstimulated from being in the city and from all of the talking to people I'd done during my poster presentation. That alone was probably 10 times as much interaction with people as I'm used to in a day, and more than 100 times the interaction I usually have with strangers. Then Ted invited everyone at the conference who had a significant association with his lab (about 16 people) to dinner at a fancy Italian place on restaurant row. I'm not one to decline a good meal, but I was feeling tired and introverted. The dinner was fantastic, but it lasted three and a half hours! By the time it ended I felt like I was about to suffer a total mental collapse. I was really glad to get back to my room and sleep.
Today I go to the conference until about 3, and then leave for the airport. Next time most of you hear from me I'll probably be in Seattle. Until then!